DESCRIPTION: (verbatim from applicant's abstract) My aim is to define the fundamental chemical principles of metal interactions with nucleic acids and nucleotides. Recently we have made observations that greatly alter important concepts about cisplatin-DNA adducts with guanines cross-linked by an N7-Pt-N7 motif. This valuable anticancer therapeutic agent, used clinically for treating cancers with growing incidence, is predicted to have continued importance well into the future. In the generally accepted hypothesis, interaction of proteins (including enzymes) with the distorted PtDNA initiates the stream of biological events causing cancer cell death. The large number of proteins found to interact with the lesion has created a lack of consensus on the biological mechanism that plagues this crucial field. Also, the world-wide effort to find a superior drug has failed, despite the testing of over 3,000 analogues differing in the carrier ligands. I hypothesize that both serious biomedical problems result from an inadequate understanding of the cisplatin adduct chemistry; this chemistry is intractable because the drug's simplicity affords few handles for elucidating the chemistry. I also hypothesize that very large and rapid dynamic motions centered at the Pt of DNA GN7-Pt-GN7 cross-link lesions occur but have not been appreciated fully by investigators. We test these hypotheses through a deliberate unique retro-modeling strategy using isomerically pure complexes with carrier ligands designed to provide the needed handles and to decrease adduct motion. We have slowed the dynamic processes by a billion-fold, thereby uncovering numerous novel N7-Pt-N7 cross-link properties undetectable by traditional approaches to Pt drug chemistry. Many of these properties could not have been foreseen, and they raise hypotheses about the PtDNA chemistry that might bestow on cisplatin its remarkable activity. I propose to advance this retro-modeling chemistry to test several new hypotheses relating to DNA duplex adducts, including (a) the existence and possible importance of single-stranded regions and novel lesion conformers, (b) the formation of cross-links, and (c) the influence of carrier ligand NH groups on structure and dynamics. Our approach will yield a new class of synthetic cisplatin analogs acting as biosensors; these biosensors will bear an identifiable imprint of adduct conformation in DNA polymers and have advantages in spectroscopic studies. Also, we will construct stable less fluxional Pt-oligonucleotide duplexes, which will be available as tools for probing biological mechanisms. Our unprecedented findings have significance beyond the field of anticancer drugs since we are identifying novel nucleic acid structures and the spectroscopic signatures such unusual structures possess.